Fundamental and Applied Wide Bandgap Semiconductor Device Modeling for Next Generation Naval Applications
Abstract
The work proposed here seeks to develop physics-based and behavioral models of wide bandgap semiconductors for applications in power conversion circuits relevant to electric navy ships. The specific semiconductors of interest are gallium nitride (GaN) and silicon carbide (SiC). The proposed physics-based models are to be developed using TCAD software as well analytical equations. These models will predict static current-voltage as well as capacitance-voltage characteristics of the various devices of interest. Physics-based models of GaN transistors based on vertical architectures are of particular interest, especially as the voltage-handling capability of GaN devices continues to increase. Model validation of the proposed physics-based models will take place through comparison of the model’s projected results with measured static behavior. Such models will enable performance projection of the high voltage capability of future GaN devices. Behavioral models will be developed using circuit simulation software to predict transient and conduction behavior of both GaN and SiC devices in power conversion circuits relevant to the navy. In particular, optimized circuit configurations and gate-driving techniques will be prescribed such that the need to trade device performance for device safety is obviated. This will be particularly advantageous since the fast switching capability of wide bandgap semiconductors, though generally a beneficial feature, can become exacerbated in the presence of significant parasitic inductance. When disrupted by significant parasitic inductance, the fast switching ability of wide bandgap devices can in turn can lead to detrimental transient behavior such as high overshoot, ringing, false turn-on, and self-sustained oscillation. The simulation models proposed here will not only be capable of predicting these anomalous behaviors, but also will lead to design considerations that can mitigate them, with minimal cost to device performance. The optimal results projected by the developed models will be experimentally validated with hardware testbeds of double pulse tests as well as DC to DC power converter circuits. Furthermore, the ability to integrate physics-based models into circuit simulation is of particular interest to the navy. Therefore, investigations for developing streamlined pathways that integrate physics-based models into circuit simulation will be undertaken. This will be useful not only for projecting the performance of existing fabricated devices, but also for projecting the performance of future theoretical high voltage devices. This will enable design engineers to determine if a theoretical device can meet specified system-level design criteria such as power density or efficiency. Where deficiencies in the projected results may exist, design engineers can work closely with device modelers to modify physics-based device parameters until the desired outcome is obtained. This will potentially reduce costs for fabricated devices, as the device design will be optimal for its intended application. Such a framework will also be useful for projecting the performance of future materials such as diamond or gallium oxide, which have yet to become technologically ready enough for circuit implementation.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 27, 2018
- Source ID
- N000141812676
Entities
People
- Raghav Khanna
Organizations
- Office of Naval Research
- United States Navy
- University of Toledo